Communicating weather data when the stakes are high
by Zachary James
The Challenge
“What is the weather in New York City?” A quick search on the internet will give you the current conditions, such as temperature, humidity, and if the sun is out. But is the information tailored to where you are located at the moment? New York is a large city, occupying more than three hundred square miles. It is also climatically diverse, with residents living in dense urban areas, on coastal islands, and everywhere in between. On a given day, two neighborhoods can experience significantly different conditions. How can we take this into account and accurately report the weather?
This is a serious question. Consider the remnants of Hurricane Ida which, in September 2021, battered the Northeast with high winds, lightning, and most concerningly, heavy rain. The National Weather Service (NWS) issued its first-ever flash flood emergency for the city and gave dire rainfall predictions. Yet New Yorkers were unsure how the rains would affect them and what the conditions would look like in their neighborhood. The result was deadly: 13 people were killed in New York City, most of whom became trapped in flooded underground apartments. It has become clear that more work is needed to communicate dangerous weather conditions to city residents.
The Project
One of the agencies that tackles challenges like this is New York City Emergency Management (NYCEM), which I worked with in the summer of 2023 as a Siegel Family Endowment PiTech PhD Impact Fellow. NYCEM is a city agency tasked with preparing for, responding to, and recovering after disasters. This is a large mandate and includes everything from dealing with terrorist attacks to coping with the arrival of a large number of asylum seekers. Yet a constant threat facing the city is that of hazardous weather.
At NYCEM, the Watch Command team monitors and reports weather conditions using forecast data and information from the National Weather Service. The agency also conducts telephone consultations with the NWS for more significant storms or widespread hazardous weather. The NWS publishes a wealth of information and data. However, it is often scattered across multiple websites and platforms, making forecast interpretation difficult and time-consuming. The Watch Command also disseminates call notes to hundreds of partners after consultations. Currently, data collection and dissemination is a laborious, manual process.
Working alongside the NYCEM meteorologist and developers, I designed an online dashboard to help address these concerns. The dashboard uses data from the National Oceanic and Atmospheric Administration (NOAA), accessed through an application program interface (API). It provides information on every neighborhood through an interactive map. Users can hover over a location on the map with the cursor to see current conditions. They can also use a slider to view future conditions. The dashboard presents all NWS watches and warnings in an easy-to-understand format.
The dashboard serves two purposes. First, it automatically consolidates data across multiple NWS sources, providing a sole-source database for information. Second, it provides decision-grade information enabling more efficient and effective decision-making by city planners, first responders, and officials based on real-time weather.
I created the weather dashboard in R, a programming language designed for statistical analyses. The Shiny software from Posit Inc. was used to help create the website. The website is currently hosted with tools from Amazon Web Services (AWS). All code for the project can be found here.
Impact and Path Forward
My work on the dashboard is only a first step. Additional work is needed so the website can handle a large number of visitors during an emergency. More data sources can be included to provide more detailed forecasts. Finally, the dashboard needs to be integrated across NYCEM services so other city agencies can quickly access it. When fully completed, it will help New York City better respond to hazardous weather.